|Home | About | Journals | Submit | Contact Us | Français|
This study explores the current patterns of reproductive health service use among young women in the USA and the changing influence of socio-demographic factors on the types of services used over time.
The study population, drawn from the two last cycles of the National Survey of Family Growth, consists of women aged 15–24 (n = 2543 in 1995, n = 2157 in 2002). We examined trends in use of ‘contraceptive services’ and ‘other reproductive health services for preventive care’ and tested for changes in the patterns of use of these services over time. Logistic regression models were used to further clarify the factors associated with the use of the two types of services in 2002.
Results show no difference in the overall use of reproductive health services in the past year but did reveal changes in the type of service sought. Use of services for contraception increased by 10 percentage points (39.3% in 1995 to 49.7% in 2002, P < 0.001), although the use of other services remained stable (53.2% in 1995, 50.2% in 2002, P = 0.14). The patterns of use varied over time, exhibiting growing social disparities. In 2002, the use of contraceptive services depended on women's age, number of partners, personal and mother's level of education, and menstrual problems. The use of other reproductive health services for preventive care varied across women's socio-economic background.
This study demonstrates increasing social differentials in the use of reproductive health services for preventive care among young women in the USA between 1995 and 2002, a finding which calls for careful monitoring in the context of limited resources.
Between 1995 and 2002, the teen pregnancy rate in the USA fell significantly, establishing a record low rate for the 1976–2002 period (Ventura et al., 2006). Most teen pregnancies in the USA are unintended and due to a combination of contraceptive nonuse and contraceptive failure (Manlove et al., 2004). The association between poor contraceptive use and teen pregnancy is supported by studies demonstrating that improved use of contraception played a key role in reducing teen pregnancy risk during the 1990s (Santelli et al., 2007). However, despite the encouraging trend toward fewer teen pregnancies, a summary of health services used by teen women published by the National Campaign to Prevent Teen Pregnancy reveals that over half of women aged 15–19 did not access any reproductive health care in the year preceding the 2002 survey (Suellentrop et al., 2008). If these women are at risk for pregnancy, it follows that there is an unmet need for reproductive health care among young women in the USA.
A recent analysis of sexual and reproductive health service use among US women of reproductive age (15–44) shows that, whereas overall use of sexual and reproductive health care services remained constant between 1995 and 2002, use of contraceptive services sharply rose over the same period of time, especially among the youngest women (Frost, 2008). On the basis of these results, we seek to explore how socio-demographic factors are associated with preventive reproductive health care service use in the USA and to understand the changing influence of such factors over time. We expand our population to include women less than 25 years old, as they account for half of all abortions performed in the USA (Guttmacher Institute, 2008) and half of all sexually transmitted infections (Weinstock et al., 2004); the proportion of pregnancies that are unintended is also highest in this age group (Finer and Henshaw, 2006). We control for sexual experience in this analysis, with the understanding that not all young women have the same reproductive health needs.
Our study also looks at factors associated with the types of reproductive health services obtained by young women. Results of the 1995 and 2002 National Survey of Family Growth (NSFG) surveys indicate that publicly-funded family planning clinics are more likely to provide contraceptive counseling and STD testing and treatment to women than private providers are (Frost, 2001, 2008). However, between 1995 and 2002, there was a decrease in the number of publicly-funded clinics receiving Medicare funding. In 2003, fewer family planning clinics provided free reproductive health care services to adolescents than in 1999 (Lindberg et al., 2006). Although the data suggest an overall increase in contraceptive service use over this period, women across the USA may not have equally participated in this increase in service use. Thus, in this context of resource vulnerability and reduced provision of family planning services to the poorest teens, we analyze the distribution of reproductive health services for preventive care obtained by women aged 15–24 who participated in the 2002 NSFG survey.
The data for this study were obtained from the 1995 and 2002 cycles of the NSFG, a population-based survey conducted by the Centers for Disease Control and Prevention's National Center for Health Statistics to provide national data on contraception, infertility, sexuality and health of men and women in the USA (Lepkowski et al., 2006). For both the 1995 and 2002 cycles, data were collected from an area probability sample representing the household population of the USA, 15–44 years of age. In-person interviews were conducted in the homes of 10 847 women in 1995 and 7643 women in 2002, with over-sampling of Black and Hispanic women. The response rate was 80%. The interviews were voluntary and confidential. The study was approved by the Institutional Review Panel for Human Subjects at Princeton University.
For the purposes of this study, we restricted the population analyzed to women between the ages of 15 and 24. We analyzed data only from women who were not currently pregnant and who had not received prenatal or post-natal care in the past year, as we hypothesized that reproductive health service use by women who were either pregnant or had recently given birth would be disproportionately greater than that of the general population. With these restrictions, our study population was 2523 in 1995 and 2157 in 2002.
In the first part of the analysis, we compare 1995 and 2002 overall reproductive health service use in the last 12 months for all women 15–24. We then stratify the results by sexual activity status (‘ever had sex’ and ‘never had sex’) and further explore changes over time by focusing on the use of reproductive health services for preventive care, which we divided into two categories: ‘contraceptive services’ and ‘other preventive reproductive health services.’ Contraceptive services include consultation, prescription and/or counseling for birth control (including sterilization). In addition, women in the 2002 survey were also asked if they had received a prescription or counseling for emergency contraception. Only 11 of the 156 women who received emergency contraception services did not report other contraceptive service use. These women represent 0.3% of the total population receiving contraceptive services. Other reproductive health services for preventive care include pap smears, pelvic exams and STI testing or treatment (excluding HIV testing and treatment). The NSFG questionnaire did not allow for the distinction between STI treatment and STI testing. It was not possible to determine the category of reproductive health services used by 73 women in 1995: 41 of them only described having received an HIV test (which may have been performed in a non-reproductive health care setting) and 32 denied receiving any of the 13 types of reproductive health services examined in the 1995 survey. In addition, 52 women described having used abortion services only (n = 6), or having only received a pregnancy test (n = 48). These women as well as the 73 described above were considered to have received services other than preventive care services.
In the second part of the analysis, we compare the determinants of contraceptive service use and other reproductive health preventive care service use in 1995 and 2002, stratifying by sexual activity status (ever had sex and never had sex). We test for changes in the patterns of health care use over time. Finally, as we identified different determinants of preventive care use in 1995 and 2002, indicating that patterns of health care use changed over time, we present a multivariate analysis for determinants of the use of the two types of preventive reproductive health services (contraceptive and other services) for the most recent year 2002. The multivariate models included all variables for which P-values were 0.25 or lower in the univariate analysis. All analyses were performed using logistic regression models. Changes in the patterns of health care use were tested by introducing an interaction term between the survey year in the factors associated with the use of services. The analyses were performed using Stata 10 SE (Stata Corporation, College Station, TX, USA). Data are weighted to account for the complex sampling design of the NSFG surveys (unequal probabilities of being included in the survey). The numbers shown in the tables are unweighted, but the percentages are weighted.
We found no difference in the overall use of reproductive health services in the preceding calendar year between 1995 and 2002, nor in the use of these services for preventive care only (Table I) In both survey years, most women used a combination of preventive care services during the 12 months preceding the survey. Thus, among women who received at least one preventive care service, 11% received contraceptive services alone, 23% received other preventive care services alone and 66% received both. When stratifying use of reproductive health services by women's sexual activity status, we found a 4.8% point increase in the use of preventive care services (including contraceptive services, STI testing/treatment, pap smear and pelvic exam) among women who had ever had sex. Results also indicate that between 1995 and 2002, women reported differences in the type of preventive services sought. There was a 10% point increase in use of contraceptive services, although there was no difference in other reproductive health service use for preventive care between 1995 and 2002 (Table II). More women received contraception (44% versus 37%, P = 0.003), contraceptive counseling (27% versus 20%, P < 0.0001), pelvic exams (61% versus 55%, P = 0.003) and testing or treatment for STIs (16% versus 10%, P < 0.0001) in 2002 than in 1995. Conversely, the same proportion reported having had a pap smear (47% in 1995 and 2002, P = 0.97). Among women who had never been sexually active, 13.7% reported having had a pap smear in the preceding 12 months (13.4% in 1995 and 13.9% in 2002) and 1.4% were tested/treated for STIs (0.6% in 1995 and 2.2% in 2002).
The increase in service use for contraception between 1995 and 2002 has not benefited all women equally (Table III). For sexually active women, those who were less educated, foreign born, uninsured at some point during the year, or daughters of less educated mothers were less likely to have increased their use of services for contraception in 2002. Conversely, the youngest women experienced a greater increase in their use of contraceptive services than did their older counterparts. Among women who had never had sex, the only difference in the patterns of contraceptive service use by year of survey pertained to the area of residence: residents from suburban metropolitan areas were less likely to exhibit an increase in family planning service use in 2002 than were others.
Although there was no overall increase in the use of other reproductive health services for preventive care between 1995 and 2002, there were changes in the patterns of use of these services over this period (Table IV). Results show a decrease in use of these other preventive care services among socially-disadvantaged young women who had previously had sex, including those who were foreign born, unemployed or not in the labor force/school, uninsured at some point during the previous year, or had less educated mothers. Conversely, there tended to be an increase in the use of these services among the youngest age group of women. Among women who had never had sex, there was a significant decrease in (non-contraceptive) preventive service use among women who were less educated, unemployed or not in the labor force/school and who reported having menstruation or ovulation problems.
In the multivariate analysis of determinants of contraceptive service use in 2002, we found that use of these services by women who were sexually experienced depended on age, number of partners, country of birth, personal level of education, mother's level of education and problems with menstruation or ovulation (Table V). Only the last factor was associated with contraceptive service use among women who had never had sex.
The second multivariate model analyzing factors associated with other preventive reproductive health service use in 2002 shows that the use of these services among those who were sexually experienced varied according to women's socioeconomic background. In particular, women not in the labor force/school were less likely to report using these services in the previous 12 months. The same was true for women who were less educated or had less educated mothers. Conversely, women with at least one sexual partner in the last 12 months and black women were more likely to have sought these services in the last year. Among women who had never had sex older age, higher level of education, and reporting difficulty with menstruation or ovulation were all predictive of service use (Table VI).
This study shows that between 1995 and 2002, the overall stability of reproductive health service use among young women in the USA conceals widening socioeconomic disparities in use of these services. Although the study shows a 10% point average increase in the use of services for contraception, young women of low socioeconomic status enjoyed smaller increases in use of these services. These growing social differentials are even more striking when it comes to other reproductive health services for preventive care (pelvic exams, pap smears, STD testing/treatment). Use of these services has remained stable on average; however, it has sharply declined among the socially disadvantaged, including women who were foreign born and lacking insurance coverage. In a recent analysis of reproductive health service use by type of provider in the USA between 1995 and 2002, Jennifer Frost shows that the rise in contraceptive service use among women of reproductive age (15–44) is almost entirely due to an increase in contraceptive care received by women visiting private doctors. In contrast, she shows the delivery of contraceptive services by other providers has remained unchanged (Frost, 2008). These findings may partly explain why contraceptive service use has mostly improved among the most advantaged young women but has stagnated among others.
Our results suggest that disparate use of reproductive services for preventive care is rooted in social inequalities, perhaps more pronounced in 2002 than in 1995. These results are consistent with the conclusions of a recent report expressing concern over the lack of improvement in contraceptive use among poor women in the USA. This plateau comes after a period of continuous improvement in contraceptive use following the passage of Title X of the Public Health Service Act in 1970, legislation enacted to provide publicly subsidized family planning (Boonstra, 2008). There was a 4% increase in contraceptive non-use among women at risk of an unintended pregnancy between 1995 and 2002 (Chandra et al., 2005), and a larger increase among women with lower education and lower income (Boonstra, 2008). In a representative sample of US women at risk for unintended pregnancy interviewed in 2004, Frost and colleagues found that less educated women were only a third as likely to use any form of contraception as those who had a college degree (Frost et al., 2007). They also found a greater risk of not using contraception among women who reported having no healthcare provider. Such disparities ultimately widen the socioeconomic gap among women experiencing unintended pregnancy. In a similar vein, Finer and Henshaw show that between 1994 and 2001, the rate of unintended pregnancy declined among adolescents, college graduates and the wealthiest women but increased among poor and less educated women (Finer and Henshaw, 2006). Interestingly, we found that lack of health insurance, often cited as a culprit in deterring use of prescription contraceptive methods, was not significantly associated with contraceptive service use, after controlling for other socio-economic factors (Culwell and Feinglass, 2007). Improving reproductive healthcare coverage for women in the USA is an essential step towards reducing social inequalities (Ranji et al., 2007), but this alone is unlikely to close the gap. Indeed, as Boonstra argues in an analysis of government program impact on reproductive health inequalities, targeted initiatives have limited impact in solving the multidimensional cause of these disparities.
In contrast to many studies, our results show greater use of contraceptive services among the youngest women. These results indicate that sexual activity accounts for much of the reported difference in healthcare utilization among teens. Therefore, this study suggests improvement in sexually active teens’ use of reproductive health care services over this period. This finding is also consistent with the increase in contraceptive use and the decline in unintended pregnancy rates among teens between 1994 and 2001 (Finer and Henshaw, 2006; Santelli et al., 2007).
Finally, we found no significant differences in the use of other reproductive health services for preventive care by area of residence. These results however, need further investigation to take into account the role of contextual factors (such as the availability of publicly-funded reproductive health services) on the use of reproductive health services in the USA.
This study reveals growing socio-economic disparities in the use of reproductive healthcare services for preventive care among young women in the USA between 1995 and 2002. In a context of resource vulnerability, it is essential to recognize the nature and extent of these disparities, and promote reproductive health policies aimed at addressing these social inequalities, which are likely to expand in the current era of economic unrest.
J.P. and C.M. were involved in designing the study, analyzing the data and wrote the article. J.T. was involved in analyzing the data and writing the article. All the authors of the paper approved the current version of the manuscript.